Hi, I clone your code and compile it in ubuntu 16.04 and g++ (Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609, but there are some errors as follow. Why?

Handheld-Multi-Frame-Super-Resolution-master/src/main.cpp:1:0:
Handheld-Multi-Frame-Super-Resolution-master/src/kernels.h: In function ‘cv::Mat getKernelWeights(int, int, T, T, T, T)’:
Handheld-Multi-Frame-Super-Resolution-master/src/kernels.h:19:25: error: expected primary-expression before ‘>’ token
kernel.ptr(i)[j] = std::exp(-0.5* (axx + (b + c)xy + d * yy));
^
Handheld-Multi-Frame-Super-Resolution-master/src/kernels.h:19:31: error: invalid types ‘int[int]’ for array subscript
kernel.ptr(i)[j] = std::exp(-0.5 (axx + (b + c)xy + d * y*y));
^
make[2]: *** [CMakeFiles/HMF.dir/main.cpp.o] Error 1
make[1]: *** [CMakeFiles/HMF.dir/all] Error 2
make: *** [all] Error 2

Can you provide more implements of handheld multi-frame super-resolutio?

hi，in Fig.8 of the paper, it seems lamda1/lamda2 is in range of 0~1, however, we know lamda1 is the dominant eigenvalue and should be greater than lamda2. I am confused, have you found this contradictory?

Hi,
I saw in another issue you mentioned that there are some details missing that is blocking this project. I found someone who implemented the same paper with some fixes in cuda and c#.
https://github.com/kunzmi/ImageStackAlignator
Maybe you can use it as reference to complete yours.